INDEX

Explanations

programming and mathematical expressions within code or structured data formats

oai_token-act-pair · gpt-4o-mini Triggered by @bot

diagrams with bend or related to code

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TagMode

-0.79

 Signalez

-0.62

bootstrapcdn

-0.61

OGND

-0.60

AddTagHelper

-0.59

ItemBackground

-0.57

UserScript

-0.55

 Numerade

-0.55

EDEFAULT

-0.55

Rujuakan

-0.54

POSITIVE LOGITS

amitié

0.35

appartamento

0.34

 alike

0.34

 constexpr

0.34

 arrivant

0.33

 secreto

0.31

entraînement

0.31

 rahasia

0.31

 sekret

0.31

 Pública

0.31

Activations Density 0.066%

programming and mathematical expressions within code or structured data formats

diagrams with bend or related to code

No Comments

No Known Activations